466 ALASKAN GLACIER STUDIES along with the paddles while a little water remained and, after that, dragging it easily over the bare, slippery surface of the fine glacial sediment. WindrBlawn Glacial Deposits on the Delta. In a number of places sand (Junes rise above the Copper River delta to a considerable height. The material is sand formerly carried out by glacial streams. There are large sand dunes 80 to 40 feet high on Long Island, just east of the Round Island Channel, which necessitate sand sheds along the railway. There is another large area of sand dunes in the 150 foot hill visible to the south of Long Island. There are also sand dunes east of Hotcake Channel, as well as on Egg Islands and other sand bars to the eastward. The latter have been built up by the surf hi the shallow water off the delta front. From Softuk Bar, near Katalla, to the mouth of Orca Inlet, these sandbars form an incomplete barrier 5 to 30 feet high and 4 to 7 miles off the front of the Copper River delta. While most of this sand comes from the Copper River and adjacent streams, some is doubtless supplied by the glacial streams from the Bering Glacier in Controller Bay and is moved along shore by the westward-setting currents. Although a normal delta in many respects, and with normal deposition by water, modified by the rise and fall of the tide, by alongshore currents, and by wind work, the Copper River delta is conspicuous as the terminus of a great area of outwash deposits, brought by the great volume of sediments from the present-day and former glaciers of this region. The Driftless Area Near Alaganik. An unforeseen feature in connection with the former expansion of the glaciers near the Copper River delta and the erosion accomplished by them, is the absence of any indication of glaciation along parts of the mountain front between Sheridan Glacier and the Copper River. At several points along the railway, as near Alaganik, there are rock pinnacles standing out from the mountain front and resembling stacks along a sea shore. Such forms as these are easily overturned by a glacier, and it seems quite probable that they are stacks produced by the waves of the Pacific before the Copper River delta was built. They cannot be considered shore forms produced after the retreat of the glacier, for there is no explanation of the failure of the waves to remove the evidences of glaciation at other points on the coast and higher above sea level than this. The conditions in the glaciated areas adjacent to the driftless region are unmistakable. That the ice expanded for some distance outside the mountain front in the region west of this driftless area is shown by the fact that the lower rock ledges in the vicinity of Sheridan Glacier bear striae, and there are many roches moutonnee rocks, like the one at the front of Sheridan Glacier: there are extensive cirques in the mountains west of Scott Glacier, where there is a hanging valley half way between Scott Glacier and the railway; the valley containing Eyak Lake, between Scott Glacier and Cordova, was much deepened and had its lower slopes steepened by the expanded southern tongue of Shephard Glacier, which filled the whole valley of Eyak Lake; along the railway, striae on the rocky spur projecting into Eyak Lake from the south show that the ice tongue split on this point, sending one glacial distributary southwestward past Cordova, and the other southeastward to join the expanded Scott Glacier. All of the glacial drift between Eyak Lake and the driftless area at Alaganik is outwash, except for the eastern terminal moraine of Sheridan Glacier which contains till and abundant granite bowlders.